Voriconazole vs. amphotericin B for neutropenic fever
PubMed
Clinical Question
For neutropenic fever refractory to antibiotics in adults treated for malignancy, is voriconazole noninferior to amphotericin in achieving clinical response?
Bottom Line
Voriconazole is a suitable alternative to amphotericin for empiric antifungal therapy in adults treated for malignancy with persistent neutropenic fever.
Major Points
For neutropenic fever in oncology patients, empiric antibiotics improves mortality compared to waiting for microbiological data before treatment. However, corollary data to support empiric antifungal therapy is much more limited. Further, for febrile neutropenia refractory to antibiotics, no trial has demonstrated superior efficacy of any particular antifungal agent.
In this 2002 trial, the NIAID Mycoses Study Group compared the efficacy and tolerability of voriconazole as a possible alternative to liposomal amphotericin B. While the trial did not meet its pre-specified noninferiority target and did not receive FDA approval for this indication, the difference between the two groups' primary outcomes did not reach statistical significance. Treatment success and mortality had a nonsignificant trend towards inferiority, however voriconazole had significantly decreased documented breakthrough fungal infections, driven by its large effect size in high-risk patients. In the other 66% of patients, amphotericin significantly outperformed voriconazole. While this suggests that triazoles may have some role in oncologic patients, it was not more well-tolerated than amphotericin.
Antifungal prophylaxis has been shown in multiple meta-analyses to reduce mortality from invasive fungal infections (IFI), in patients with >6% risk of IFI and receiving HSCT or other treatments expected to cause profound, persistent neutropenia (<100 PMNs for over 7 days). This has not been demonstrated in patients receiving more typical chemotherapy regimens.
Guidelines
NCCN Guidelines: Prevention and Treatment of Cancer-Related Infections (2016)
- Voriconazole is an option (category 2B) for empiric therapy in patients at high risk for invasive mold infection.
ASCO Clinical Practice Guidelines (2013)
- Only use antibacterial and antifungal prophylaxis if neutrophils are expected to remain < 100/μL for >7 days, unless high risk for complications or mortality (patient characteristics, type of cancer, type of treatment)
- Oral triazole is preferred for antifungal prophylaxis. We recommend their use in patients with >10% risk of invasive Candida infections or >6% risk of Aspergillosis.
IDSA Clinical Practice Guidelines (2010)
- For persistent neutropenic fever ( >4 days), recommend obtaining clinical, microbiology, or radiographic evidence of invasive fungal infection before initiation anti-fungal treatment.
- When treating, there is no clear evidence to distinguish efficacy of amphotericin, itraconazole, voriconazole, or caspofungin.
Design
- Multicenter randomized, controlled trial (open label)
- N=837
- Voriconazole (n=415), 3 mg/kg IV q12h after 6mg/kg IV q12h x2 loading dose, with protocol-based adjustments as below
- Amphotericin B (n=422), liposomal, 3 mg/kg per day, with protocol-based adjustments as below
- Setting: 73 centers in the United States, Canada, France, India, Italy, and the United Kingdom
- Enrollment: March 1998 - September 1999
- Median follow-up: not stated
- Analysis: Intention-to-treat, noninferiority defined as < 10 percentage points difference
- Primary outcome: Treatment success, defined below
Population
Inclusion Criteria
- Age ≥12 years
- Received chemotherapy or HSC transplantation
- Neutropenia (ANC<500 for 4 days and <250 for 1 days)
- Febrile despite 96 hrs of systemic antibacterial therapy
Exclusion Criteria
- Documented invasive fungal infection at the time of randomization
- AST, ALT, TB, or Alk Phos >5x upper limit of normal
Baseline Characteristics
- Differences between groups appear grossly insignificant, though p-value not calculated
- Demographics: Age 45-46, 51-56% male, 78% Caucasian
- Cancer: New leukemia (31%), relapsed leukemia (21%), lymphoma (14%), solid tumor (13%), multiple myeloma (5.5%), other (15%)
- Treatment type: stem cell transplant (48-51%)
- Disease extent: 34% high risk, 66% moderate risk.
Interventions
Patients were randomized to:
- Voriconazole (n=415), 3 mg/kg IV q12h after 6mg/kg IV q12h x2 loading dose, with protocol-based options to switch to 4 mg/kg IV q12h or, after 3 days, to switch to 200-300 mg PO q12h
- Amphotericin B (n=422), liposomal, 3 mg/kg per day, with protocol-based options to switch to 6 mg/kg or, if intolerant, 1.5 mg/kg
The primary outcome, treatment success, required meeting all of the following: surviving for 7 days beyond end of therapy, remaining afebrile, with clinical suspicion of successfully treated fungal infection, without premature treatment discontinuation or breakthrough fungal infection.
Non-inferiority was defined as being able to exclude with 95% confidence a 10 percentage point decrease in the primary outcome
Subgroup analysis was performed on high risk patients, defined as those with allogeneic transplants or relapsed leukemia
Outcomes
Comparisons are voriconazole vs. amphotericin
Primary Outcomes
- Treatment success
- Overall: 26.0% vs 30.6%, or - 4.6 percentage points (95% CI -10.6 to +1.6; P<0.05)
Secondary Outcomes
- Survival at Day 7
- 92.0% vs 94.1 %, or -2.0 percentage points (95% CI -5.5 to +1.4; P<0.05)
- Breakthrough fungal infections within 7 days of end of therapy
- 1.9% vs 5.0 %, or + 3.1 percentage points (95% CI +0.6 to +5.5; P<0.05)
- Discontinuation rate
- 9.9 % vs 6.6%, +3.3% (95% CI -0.5 to +7.0; P<0.05)
Subgroup analysis
- Among high-risk patients
- Treatment Success: 32.0% vs 30.0%, or + 2.0 percentage points (95% CI -9.0 to +12.4; P<0.05)
- Breakthrough fungal infections: 1.4% vs 9.2%, or + 7.8 percentage points (P=0.003)
- Among moderate-risk patients
- Treatment Success: 23.0% vs 31.0%, or - 8.0 percentage points (95% CI -15.2 to -0.4; P<0.05)
- Death due to progressive cancer: 3.1% vs 1.1% (P=0.06)
Adverse Events
Abnormal vision: 21.9% vs 0.7%; P<0.001
Pain: 0.2% vs 1.9-4.0%; P<0.001 (Chest, Abdomen, Back, Flank)
Dyspnea: 0.7% vs 8.8%; P<0.001
AKI: 10.4% vs 19.0%; P<0.001, however insignificant for Cr>2x baseline (7.0 vs 7.6%)
Hypokalemia: 16.4% vs 31.0%; P<0.001
Criticisms
- The study was underpowered to rule out a difference in overall mortality, which had a trend toward inferiority (2.1% 7-day mortality difference) in the voriconazole group
- Additional mortality data from IFI may not have been captured by only examining 7 days out
- Treatment success is a difficult primary outcome to capture, although this particular composite outcome appears well-reasoned and pre-defined
- The subgroup analysis suggested the a large difference in treatment effect in those with and without relapsing leukemia or allogeneic transplant, and may have masked significant differences in both groups
- A third group without antifungal treatment may have offered insight into whether empiric antifungal therapy improves mortality in the study population
Funding
Funded by NIH (NIAID) and Pfizer (producer of Vfend/voriconazole)
Further Reading
NCCN Guidelines: Prevention and Treatment of Cancer-Related Infections. https://www.nccn.org/professionals/physician_gls/pdf/infections.pdf
ASCO Guidelines: Antimicrobial Prophylaxis and Outpatient Management of Fever and Neutropenia in Adults Treated for Malignancy. http://jco.ascopubs.org/content/31/6/794.full
IDSA Guidelines: Clinical Practice Guideline for the Use of Antimicrobial Agents in Neutropenic Patients with Cancer. http://cid.oxfordjournals.org/content/52/4/e56.full
Voriconazole Compared with Liposomal Amphotericin B for Empirical Antifungal Therapy in Patients with Neutropenia and Persistent Fever. http://www.nejm.org/doi/full/10.1056/NEJM200201243460403